(22)

(22)

(26)

(27)
(28)

(29)

(30)
(32)

(32)

(33)

DAWBER, T.R., KANNEL, W.B., GORDON, T. Coffee and cardiovascular
disease. Observations from the Framingham Study. New England Journal of
Medicine 291(17): 871-874, October 24, 1974.

DUNN, W.L., JR. Experimental methods and conceptual models as applied to
the study of motivation in cigarette smoking. In: Dunn, W.L., Jr. (Editor).
Smoking Behavior: Motives and Incentives. Washington, D.C., V.H. Winston
and Sons, 1973, pp. 93-111.

EISINGER, R.A. Psychosocial predictors of smoking behavior change. Social
Science and Medicine 6(1): 187-144, February 1972.

EISINGER, R.A. Psychosocial predictors of smoking recidivism. Journal of
Health and Social Behavior 12: 355-362, December 1971.

EVANS, R.R., BORGATTA, E.F.. BOHRNSTEDT, G.W. Smoking and MMPI
scores among entering freshmen. Journal of Social Psychology 73: 187-140,
October 1967.

EYSENCK, H.J. Personality and the maintenance of the smoking habit. In:
Dunn, W.L., Jr. (Editor). Smoking Behavior: Motives and Incentives.
Washington, D.C., V-H. Winston and Sons, 1973, pp. 118-146.

EYSENCK, H.J. Smoking, Health and Personality. New York, Basic Books,
1965, pp. 75-106.

EYSENCK, HJ. Smoking, personality and psychosomatic disorders. Journal of
Psychosomatic Research 7(2): 107-130, October 1968.

FOOTE, E. The time has come: Cigarette advertising must be banned. In:
Jarvik, M.E., Cullen, J.W., Gritz, E.R., Vogt, T.M., West, L.J. (Editors).
Research on Smoking Behavior. NIDA Research Monograph 17. US.
Department of Health, Education, and Welfare, Public Health Service,
Aleohol, Drug Abuse, and Mental Health Administration, National Institute
on Drug Abuse, DHEW Publication No. (ADM) 78-581, December 1977, pp.
339-346.

FOSS, R. Personality, social influence and cigarette smoking. Journal of Health
and Social Behavior 14(3): 279-286, September 1973.

GOODE, E. Cigarette smoking and drug use on a college campus. International
Journal of the Addictions 7(1): 133-140, February 1972.

GREEN, D.E. Psychological factors in smoking. In: Jarvik, M.E., Cullen, J.W.,
Gritz, E.R., Vogt, T.M., West, L.J. (Editors). Research on Smoking Behavior.
NIDA Research Monograph 17. U.S. Department of Health, Education, and
Welfare, Public Health Service, Alcohol, Drug Abuse, and Mental Health
Administration, National Institute on Drug Abuse, DHEW Publication No.
(ADM) 78-581, December 1977, pp. 149-155.

GROVES, W.E. Patterns of college student drug use and lifestyles. In:
Josephson, E., Carroll, E.B. (Editors). Drug Use: Epidemiological and
Sociological Approaches. Washington, D.C., Halsted Press, 1974, pp. 241-275.

GUILFORD, J.S. Factors Related to Successful Abstinence from Smoking:
Fina] report. U.S. Public Health Service, Division of Chronic Diseases, Bureau
of State Services, July 1966, 170 pp.

HERMAN, C.P. External and internal cues as determinants of the smoking
behavior of light and heavy smokers. Journal of Personality and Social
Psychology 30(5): 664-672, November 1974.

HJELLE, L.A., CLOUSER, R. Internal-externai control of reinforcement in
smoking behavior. Psychological Reports 26(2): 562, April 1970.

HORN, D. Some factors in smoking and its cessation. In: Borgatta, E.F., Evans,
R.R. (Editors). Smoking, Health, and Behavior. Chicago, Aldine Publishing
Company, 1968, pp. 12-21.

IKARD, F.F., GREEN, D.E., HORN, D. A scale to differentiate between types
of smoking as related to the management of affect. International Journal of
the Addictions 4(4): 649-659, December 1969.

18—27
(39)

(40)

(41)

(42)
(48)
(44)
(45)

(46)

(47)

(48)
(49)
(50)

(51)

18—28

JACOBS, M.A. The addictive personality: Prediction of success in a smoking
withdrawal program. Psychosomatic Medicine 34(1): 30-38, January/February
1972.

JACOBS, M.A., KNAPP, P.H., ANDERSON, L.S., KARUSH, N., MEISSNER,
R., RICHMAN, SJ. Relationship of oral frustration factors with heavy
cigarette smoking in males. Journal of Nervous and Mental Diseases 141(2):
161-171, August 1965.

KANDEL, D. Interpersonal influences on adolescent illegal drug use. In:
Josephson, E., Carroll, E.E. (Editors). Drug Use: Epidemiological and
Sociologicai Approaches. Washington, D.C., Halsted Press, 1974, pp. 207-240.

KANDEL, D. Stages in adolescent involvement in drug use. Science 190: 912-
914, November 1975.

KANEKAR, S., DOLKE, A.M. Smoking, extraversion, and neuroticism.
Psychological Reports 26: 384, February-June 1970.

KOZLOWSKI, L.T. Effect of caffeine on coffee drinking. Nature 264 (5584):
354-355, November 25, 1976.

KOZLOWSKI, L.T. Effects of caffeine consumption on nicotine consumption.
Psychopharmacology 47: 165-168, May 28, 1976.

KOZLOWSKI, L.T., HARFORD, M.A. On the significance of never using a
drug: An example from cigarette smoking. Journal of Abnormal Psychology
85(4): 433-434, August 1976.

LARSON, P.S., HAAG, H.B., SILVETTE, H. Tobacco: Experimental and
Clinical Studies: A Comprehensive Account of the World Literature. Balti-
more, Williams and Wilkins Company, 1961, pp. 526-551.

LARSON, P.S., SILVETTE, H. Tobacco: Experimental and Clinical Studies,
Supplement I. Baltimore, Williams and Wilkins Company, 1968, pp. 259-22.

LEBOVITS, B., OSTFELD, A. Smoking and personality: A methodological
analysis. Journal of Chronic Diseases 23(10/11): 813-821, March 1971.

LEBOVITS, B., OSTFELD, A.M., MOSES, V.K., PAUL, O. Cigarette smoking
and personality. Journal of Chronic Diseases 25(10/11):; 581-598, 1972.

LEVENTHAL, H., AVIS, N. Pleasure, addiction, and habit: Factors in verbal
report or factors in smoking behavior? Journal of Abnormal Psychology 85(5):
478-488, October 1976.

LEVITT, K.E. The television cigarette commercial: Teenage transducer or
paper tiger? Yale Scientific Magazine 45(1): 10-18, October 1970.

LICHTENSTEIN, E., KEUTZER, C.S. Further normative and correlational
data on the internal-external (I-E) control of reinforcement scale. Psychologi-
cal Reports 21(3): 1014-1016, December 1967.

MATARAZZO, J.D., SASLOW., G. Psychological and related characteristics of
smokers and nonsmokers. Psychological Bulletin 57(6): 493-513, November
1960.

MAUSNER, B. An ecological view of cigarette smoking. Journal of Abnormal
Psychology 81(2): 115-126, Aprit 1973.

MCARTHUR, C., WALDRON, E.. DICKINSON, J. The psychology of smoking.
Journal of Abnormai Psychology 56(2): 267-275, March 1958.

MCKENNELL, A.C. British research into smoking behavior. In: Borgatta, E.F.,
Evans, R.R. (Editors). Smoking, Health, and Behavior. Chicago, Aldine
Publishing Company, 1968, pp. 140-164.

MCKENNELL, A.C. A comparison of two smoking typologies, Research Paper
12. London, Tobacco Research Council, 1973, 95 pp.

MEYER, A.S., FRIEDMAN, L.N., LAZARSFELD, P.F. Motivational conflicts
engendered by the on-going discussion of cigarette smoking. In: Dunn, W.L.,
Jr. (Editor). Smoking Behavior: Motives and Incentives. Washington, D.C.,
V.H. Winston and Sons, 1978, pp. 243-254.
(60)

(61)

(62)
(68)
(64)

{65)

(66)

(67)

(68)

(69)

(70)

(72)

(72)

(78)
(74)

(75)

(#6)

NATIONAL CANCER INSTITUTE. AMERICAN CANCER SOCIETY. Ciga-
rette Smoking Among Teen-agers and Young Women. U.S. Department of

’ Health, Education, and Welfare, Public Health Service, National Institutes of
Health, National Cancer Institute, DHEW Publication No. (NIH) 77-1203,
1975, 31 pp.

NATIONAL CLEARINGHOUSE FOR SMOKING AND HEALTH. Adult Use
of Tobacco, 1975. U.S. Department of Health, Education, and ‘Welfare, Public
Health Service, June 1976, 23 pp.

NESBITT, P.D. Chronic smoking and emotionality. Journal of Applied Social:
Psychology 2(2): 187-196, April-June 1972.

NESBITT, P.D. Smoking, physiologieal arousal, and emotional response. Journal
of Personality and Social Psychology 25(1): 187-144, January 1973.

NISBETT, R., WILSON, T.D. Telling more than we can know: Verbal reports
on mental processes. Psychological Review 84(3): 231-259, May 1977.”

O'ROURKE, T.W. Research on smoking behavior: Some limitations and
suggestions for improvement. Public Health Reviews 2(1): 105-112, February
1973.

PALMER, A.B. Some variables contributing to the-onset of cigarette smoking
among junior high school students. Social Science and Medicine 4(3): 359-366,
September 1970.

PETO, J. Price and consumption of cigarettes: A case for intervention? Britisn
Journal of Preventive and Social Medicine 28: 241-245, 1974.

REEDER. L.G. Sociocultural factors in the etiology of smoking behavior: An
assessment. In: Jarvik, M.E., Cullen, J.W., Gritz, E.R., Vogt, T.M., West, LJ.
(Editors). Research on Smoking Behavior. NIDA Research Monograph 17. US.
Department of Health, Education, and Welfare, Public Health Service,
Alcohol, Drug Abuse, and Mental Health Administration, National Institute
on Drug Abuse, DHEW Publication No. (ADM) 78-581, December 1977, pp.
186-200. ©

REYNOLDS, C., NICHOLS, R. Personality and behavioral correlates of

_ cigarette smoking: One-year follow-up. Psychological Reports 38(1): 251-258,
February 1976. ; ,

RODE, A., ROSS, R., SHEPHARD, R.J. Smoking withdrawal programme.
Personality and cardiorespiratory fitness. Archives of Environmental Health
24: 27-36, January 1972.

RUDOLPH, J.P., BORLAND, B.L. Factors affecting the incidence and
acceptance of cigarette smoking among high school students. Adolescence
11(44): 519-525, Winter 1976.

RUSSELL, M.A.H. Changes in cigarette price and consumption by men in
Britain, 1946-71: A preliminary analysis. British Journal of Preventive and
Social Medicine 27: 1-7, 1978.

RUSSELL, M.A.H. Cigarette smoking: Natural history of a dependence
disorder. British Journal of Medical Psychology 44(1): 1-16, May 1971.

RUSSELL, M.A.H. The smoking habit and its classification. Practitioner 212:
791-800, June 1974.

RUSSELL, M.A.H. Smoking problems: An overview. In: Jarvik, M.E., Cullen,
J.W., Gritz, E.R., Vogt, T.M., West, LJ. (Editors). Research on Smoking
Behavior. NIDA Research Monograph 17. U.S. Department of Health,
Education, and Welfare, Public Health Service, Alcohol, Drug Abuse and
Mental Health Administration, National Institute on Drug Abuse, DHEW
Publication No. (ADM) 78-581, December 1977, pp. 18-34.

RUSSELL, M.A.H., PETO, J., PATEL, U.A. The classification of smoking by
factorial structure of motives. Journal of the Royal Statistical Society A 137
(Part 3): 313-332, 1974.

18—29
(77)

(80)

(81)
(82)

(83)

(34)

(83)

(36)

(87)

{88)

(39)

(90)

(41)

(92)

(93)

(94)

18—30

RYAN, FJ. Cold turkey in Greenfield, lowa: A follow-up study. In: Dunn, W.L.,
dr. (Editor). Smoking Behavior: Motives and Incentives. Washington, D.C.,
V.H. Winston and Sons, 1973, pp. 281-241.

SALBER, E.J., ABELIN, T.. Smoking behavior of Newton school children—5-
year follow-up. Pediatries 40 (8, part I): 368-372, September 1967. .

SALBER, E.J.. WELSH, B.. TAYLOR, S.V. Reasons for smoking given by
secondary school children. Journal of Health and Human Behavior 4: 118-129,
1963.

SCHACHTER, S. Nesbitt’s paradox. In: Dunn, W.L., Jr. (Editor). Smoking
Behavior: Motives and Incentives. Washington, D.C., V.H. Winston and Sons,
1978, pp. 147-155.

SCHACHTER, S. Nicotine regulation in heavy and light smokers. Journal of
Experimental Psychology: General 106(1): 5-12, March 1977.

SCHACHTER, S. Pharmacological and psychological determinants of smoking.
Annals of Interna] Medicine 88(1): 104-114, January 1978.

SCHACHTER, 8., KOZLOWSKI, L.T., SILVERSTEIN, B. Effects of urinary
pH on cigarette smoking. Journal of Experimental Psychology: General 106(1):
13-19, March 1977.

SCHACHTER, ©. SILVERSTEIN, B., PERLICK, D. Psychological and
pharmacological explanations of smoking under stress. Journal of Experimen-
tal Psychology: General 106(1): 31-40, March 1977.

SCHUBERT, D.S.P. Arousal seeking as a central factor in tobacco smoking
among college students. International Journal of Social Psychiatry 11(8): 221-
225, Summer 1965.

SCHUMAN, L.M. Patterns of smoking behavior. In: Jarvik, M.E., Cullen, J.W.,
Gritz, E.R., Vogt, T.M., West, L.J. (Editors). Research on Smoking Behavior.
NIDA Research Monograph 17. U.S. Department of Health, Education, and
Welfare, Public Health Service, Alcohol, Drug Abuse, and Mental Health
Administration, National Institute on Drug Abuse, DHEW Publication No.
(ADM) 78-581, December 1977, pp. 36-65.

SCHWARTZ, J.L. Insights into the smoker’s profile. Journal of School Health
40: 127-131, March 1970.

SCHWARTZ, J.L., DUBITZKY, M. Requisites for success in smoking withdraw-
al. In: Borgatta, E.F., Evans, R.R. (Editors). Smoking, Health, and Behavior.
Chicago, Aldine Publishing Company, 1968, pp. 231-247.

SIMON, W.E., PRIMAVERA, L.H. The personality of the cigarette smoker:
Some empirical data. International Journal of the Addictions 11(1): 81-94
February 1976.

SMITH, G.M. Personality correlates of cigarette smoking in students of college
age. Annals of the New York Academy of Sciences 142 (Article 1): 308-321,
March 15, 1967.

SMITH, G.M. Personality and smoking: A review of the empirical literature. In:
Hunt, W.A. (Editor), Learning Mechanisms in Smoking. Chicago, Aldine
Publishing Company, 1970, pp. 42-61.

SPEVACK, M., PIHL, R.O. Nonmedical drug use by high school students: A
three-year survey study. International Journal of the Addictions 11(5): 755-
792, Octuber 1976.

SROLE, L., FISCHER, A.K. Smoking behavior 1953 and 1970: The midtown
Manhattan study. In: Dunn, W.L., Jr. (Editor). Smoking Behavior: Motives
and Incentives. Washington, D.C., V.H. Winston and Sons, 1973, pp. 255-265.

STOKES, J.P. Personality traits and attitudes and their relationship to student
drug using behavior. International Journal of the Addictions 9(2): 267-287,
April 1974.

'
(95)

(96)

(97)

(98)

(99)

(100)

(101)

(102)

(103)

(104)
(108)

(106)

STRAITS, B.C. The discontinuation of cigarette smoking: A multiple discrimi-
nant analysis. In: Zagona, S.V. (Editor). Studies and Issues in Smoking
Behavior. Tuscon, University of Arizona Press, 1967, pp. 79-81.

THOMAS, C.B. The relationship of smoking and habits of nervous tension. In:
Dunn, W.L., Jr. (Editor). Smoking Behavior: Motives and Incentives.
Washington, D.C., V.H. Winston and sons, 1973, pp. 157-170.

TOMKINS, S. A modified model of smoking behavior. In: Borgatta, E.F., Evans,
R.R. (Editors). Smoking, Health, and Behavior. Chicago, Aldine Publishing
Company, 1968, pp. 165-186.

U.S. PUBLIC HEALTH SERVICE. Smoking and Health. Report of the
Advisory Committee to the Surgeon General of the Public Health Service. U.S.
Department of Health, Education, and Welfare, Public Health Service, Center
for Disease Control. PHS Bulletin No. 1103, 1964, 387 pp.

U.S. PUBLIC HEALTH SERVICE. State Legislation on Smoking and Health:
1976. U.S. Department of Health, Education, and Welfare, Public Health
Service, Center for Disease Control, Bureau of Health Education, National
Clearinghouse for Smoking and Health, December 1976, 73 pp.

WARNER, K.E. The effects of the antismoking campaign on cigarette
consumption. American Journal of Public Health 67 (7): 645-650, 1977.

WATERS, W.E. Smoking and neuroticism. British Journal of Preventive and
Social Medicine 25(3): 162-164, August 1971.

WEATHERLEY, D. Some personality correlates of the ability to stop smoking
cigarettes. Journal of Consulting Psychology 29(5): 483-485, October 1965.

WEST, D.W., GRAHAM, S., SWANSON, M., WILKINSON, G. Five year
follow-up of a smoking withdrawal clinic population. American Journal of
Public Health 67(6): 5386-544, June 1977.

WHITESIDE, T. Smoking still. New Yorker 50(39): 121-151, November 18, 1974.

WIGGINS, J.S. Personality structure. In: Farnsworth, P.R., Rosensweig, M.R.,
Polefka, J.T. Annual Review of Psychology 19. Palo Alto, California, Annual
Reviews, Inc., 1968, pp. 293-350.

WYNDER, ELL. Interrelationship of smoking to other variables and preventive
approaches. In: Jarvik, M.E., Cullen, J.W., Gritz, E.R., Vogt, T.M., West, L.J.
(Editors). Research on Smoking Behavior. NIDA Research Monograph 17. U.S.
Department of Health, Education, and Welfare, Public Health Service,
Alcohol, Drug Abuse, and Mental Health Administration, National Institute
on Drug Abuse, DHEW Publication No. (ADM) 78-581, December 1977, pp. 67-
97.

18—31
19. MODIFICATION OF SMOKING
BEHAVIOR.

 

National Institute on Drug Abuse
Contents

 

 

 

Introduction 0.0.0.0... cece cece cece eee eee nena nna es 5
Methodological Issues............0.ccce cee reeesee ee er tere ee neta tees 5
Review of General, Nonspecific Interventions ................- 9
Public Health Educational Campaigns..............:..02+ 9
Public Service and Proprietary Clinics ..............8 10
Individual and Medical Counseling .............-..ceeee 12
Large-Scale Coronary Prevention Trials ................- 14

 

Controlled Experimental Research on Intervention

 

 

Strategies .......cccccee cess reece ener eee neeeseees eee teeeeeranees 16
Drug Treatments...........:cccceeeeeee nee te encase tence ns 16
Hypm0is ..........0ceeccees eee e eee cere teen ect eer neee nner anne 7
Social Psychological Approaches ............50.:seeesee seen es 18
Social Learning and Behavior Modification

Approaches ........0...c0cce ce ceee eee ee center nee aren een eea eens 19
Self-Control Strategies...........c.ccceceeee eee ee eee ees 20
Stimulus Control ............cccccec esse eee rene eee eee 20
Contingency Contracting.............2-:-eeeee ees 21

Other Self-Control Strategies ............6: ee 22
Aversion Strategies ..........ccccceceeeeee eee ne erences 22
Electric Shock ..........::.:eceecee ee cee ener eee eens 23

Covert Sensitization .............cce ccc ee eens eect eee 23
Cigarette Smoke Aversion ...........:.:0::seeeeees 24
Satiation.........ccc cc cceee cece ese ee ee ee eens en neneeneeees 25
Medical Risks of Aversive Smoking............-. 25

Less Stressful Alternatives ........0...cc ee 26
Multicomponent Interventions ...........::.eeeeees 27
Treatment Innovations .........0.... cece cence eee eee 28
Controlled Smoking.............. 0: cece eececce seen ee eneeees 29
Maintenance of Nonsmoking...............:c0cc:eeeeeceneeees 30
General Overview of Data.........0.c ccc ccc cece e eee e rect enes 32
Status of Methodology ...........0..cccceee sent eset treet enes 32
Implications of the Data ..........::.c: eee se eee ener 33
Recommendations for Future Research .........-....:-.ee 34
Objective Measures of Smoking ...........00 cee 34

19-—3
 

Maximizing Unaided Cessation. ...............ccccececeeseeee 35
Development of Maintenance Strategies ..........0...00. 35
Evaluation of Existing Programs and Procedures ..... 35
References 2.00.0... cccsescccee ec ecneeeeeeneneneeeseeaenseeteenens 37
Introduction

Since the health consequences of smoking became more evident in the
early 1960’s, the development of techniques to aid smokers to quit have
proliferated. The methods have ranged widely from gimmicks and
over-the-counter cessation aids to formal programs and clinics (368,
376). Thus, the concerned professional or layman with an interest in
assisting smokers in the process of cessation may find it very difficult
to decide which intervention strategy is best or most useful. The social
relevance of the topic has focused much of the effort in the field
toward clinical presentations of what logically appeared to be the best
withdrawal techniques or strategies rather than toward careful
research to define what strategy, method, or program is most effective
in producing long-term successes or positive changes in smoking
behavior. Remarkably, a wide variety of interventions has been offered
and recommended to the public, but outcome data needed for critical
appraisal of them are scarce.

The task of evaluating the relative efficacy of programs and
techniques has been very adequately done in numerous past and recent
reviews (24, 26, 29, 40, 171, 200, 224, 226, 230, 245, 366, 368, 876, 418).
Therefore, this review cen be selective in order to allow discussion of
critical topics and encourage new developments in the field. The reader
is referred to the other available reviews to obtain a more detailed
discussion of topics that are here given brief treatment.

Methodological Issues

Any reviewer of the literature on strategies to modify smoking
behavior is faced with the difficult task of sorting through outcome
research that is permeated by many methodological flaws and
deficiencies (24, 26, 224, 226, 366, 368, 376). Despite the facts that
smoking behavior offers an objectively measurable target behavior,
that potential treatment participants are numerous, and that the
normal treatment context affords the opportunity for both good
internal and external validity (24, 200, 226, 393), a number of
methodological inadequacies continues to plague the field (26, 29, 226,
368, 376, 413). Therefore, the methodology and design problems that
most commonly limit the appraisal of existing outcome data will be
briefly summarized. Anyone concerned with smoking withdrawal
prograrhs or research, however, should refer to other comprehensive
evaulations of these issues presented by Bernstein (24), Schwartz (366,
376), Lichtenstein and Danaher (226), and the National Interagency
Council on Smoking and Health’s (NICSH) Guidelines for Research on
the Effectiveness of Smoking Cessation Programs (272).

The most pervasive problem in the evaluation of outcome data from
smoking cessation programs is the validity of the treatment results.
Almost all clinics and research studies have relied primarily upon

19--5
unverified self-reports of smoking as their critical dependent measure.
Unfortunately, the verbal or written requests for estimates of number
of cigarettes currently smoked per unit of time depend upon the
participant's accuracy and honesty (226), are subject to nonspecific
demand characteristics (especially during and after treatment) (226),
and appear to be highly influenced by digit-bias (that is, given in
multiples of 5 or 1/2 pack units) (423). One study collecting global
estimates under different conditions on the same day found question-
able reliability (423). Thus, studies based only on global, unverified
self-reports of smoking behavior must be viewed with skepticism.

Because of these factors, the rate measure based on such global
estimates tends to be more an ordinal than a ratio variable (396).
Nevertheless, rate-per-unit-of-time data often have been preferred
over the dichotomous abstinent-nonabstinent or percent-reduction
categories, which clearly require the use of less powerful nonparame-
tric statistical analyses (226, 393, 396). The use of self-monitoring
recording has been recommended in various forms (109, 198, 226, 250,
272) and commonly used in many studies to enhance both the reliability
and psychometric qualities of the rate data. However, the procedure is
known to be reactive (198, 250), is still susceptible to the demand
characteristics (198, 226), and tends to underestimate the “real”
baseline or follow-up rate (109, 198, 226, 250).

Studies not relying on smoking rates as the primary dependent
measure have commonly utilized various and often undefined success-
failure categories to minimize the problems of self-report data (24,
366). Standard categories have been suggested to avoid ambiguity
(272); however, the primary evaluation of treatment-results based on
abstinence data can be recommended for several reasons. First,
abstinence is the primary goal of almost all smokers seeking treatment
(24, 25, 40, 171, 226, 366). Second, follow-up data on smokers have
indicated that most smokers who fail to attain abstinence eventually
return to baseline smoking rates (24, 26, 171, 251). Third, analyses of
rate data can yield statistically significant treatment effects even with
a clinically insignificant proportion of participants abstinent at follow-
up (251, 366, 376). Fourth, abstinence reports are less susceptible to
nonspecific demand characteristics and the reactivity of self-monitor-
ing (226). Nevertheless, when derived from reliably collected self-
monitoring data, cigarettes-per-day rate data or the more -precise
percentage-or-baseline (current smoking + pretreatment smoking
rate x 100) variable (199, 200, 226) can be very helpful as secondary
measures for testing finer theoretical questions with parametric
statistical techniques (24, 200, 226, 272). Because treatment will often
produce a marked, positive skewness in the distributions of rates (that
is, greatly increased frequency of rates at or near zero), care should be
taken to test the homogeneity of variance and to apply transforma-

19—6
tions as necessarv before utilizing analysis-of-variance procedures,
especially with cell frequencies of unequal size (71, 292, 445).

Optimally, self-report data on smoking should be validated by an
objective measure. False reporting has now been documented in both
children (99, 154, 262) and adults in cessation programs (47, 82, 178,
283). Natural-environment informants or observers have been recom-
mended and used in many studies, but the systems are reactive,
difficult to maintain, and, owing to possible collusion, have question-
able validity (47, 226). Biochemical tests for objectively measuring
smoking exposure are clearly more desirable. Measurements of blood
carboxyhemoglobin (COHb) (61, 192, 320, 330, 397, 427) and thiocya-
nates (SCN-) in biologie fluids (18, 54, 75, 83, 238, 299, 300, 444) have
been demonstrated to be reliable indicators of smoking behavior.
Concentrations of carbon monoxide (CO) in alveolar air is directly
proportional to blood COHb concentrations (61, 320, 330, 397) and has
heen recommended as a simple validating tool (208). However, co
concentrations have a very short half-life (330, 397) and show high
diurnal variability (61, 258, 330). Thus, SCN concentrations that have a
biologic half-life of approximately 14 days (299) are more suited for
validation of self-reports (47, 54, 423, 424). Determinations of serum
SCN- have been more common (47, 54, 83, 423), but tests of urine or
saliva are also possible and may be more practical in many clinical
settings (18, 99, 262). Unfortunately, COHb levels are affected by
various environmental exposures (192, 397, 427) and SCN- concentra-
tions can be elevated by diet (47). Singly, however, they provide a
crude measure of smoking rate (423, 424) with adequate discrimination
between smokers and nonsmokers; together they appear to provide a
very powerful test of abstinence (423, 424).

In summary, researchers should be aware that uncorroborated self-
reports may lead to an overestimation of success, especially in
situations where subjects are under social pressure to quit or to report
quitting. The addition of objective biological assays can help to
validate self-report data and improve the ability to assess outcome,
using the self report as a low-cost, easily obtainable, dependent
measure.

In addition to the problem of questionable validity of self-reports
that faces all researchers, various design deficiencies also plague the
field (24, 200, 226, 272, 304, 366, 367, 376, 398). First, attributions of
causality of outcome results to independent treatment factors are
virtually impossible without systematic designs, including appropriate
experimental controls (24, 56, 391). Initial demonstrations of efficacy
may be evaluated relative to commonly expected norms of success (245,
304); such clinical demonstrations must then be replicated versus
appropriate control conditions, especially attention-placebo controls
(24, 26, 200, 226, 230, 245, 251, 272, 304, 366, 367, 376, 398). Few
procedures or programs developed in clinical settings have progressed

19—7
to experimental validation (24, 40, 245, 304, 366, 367, 376, 898, 418).
Moreover, Straits (398) has suggested that the strength of laboratory
research involves testing more complicated questions than treatment
efficacy. Factorial designs enable one to evaluate specific treatment
effects as well as more complex multidimensional and interactional
effects and thus permit the simultaneous testing of several theoretical
issues (398).

Systematic treatment evaluations must also include comprehensive
and adequate follow-up of participants (24, 26, 171, 272, 366, 368, 376).
Almost all treatments are able to show dramatic post-treatment
effects, but rapid relapse in most participants has been the norm (170,
171, 251, 366). Therefore, no treatment can be adequately evaluated
without long-term follow-up data. Recidivism tends to be the greatest
during the first 3 to 4 months after treatmeht and relatively slight
after 6 months (170, 171), but a 1-year follow-up remains highly
recommended (272, 366, 368, 376).

Comprehensiveness of follow-up is as important as length, if not
more so. Schwartz (366, 368, 376) has strongly emphasized that all
participants, including early-treatment dropouts, should be used in
computing treatment effectiveness. Additional analyses of subjects
completing most treatments are useful to clarify theoretical issues (24,
226); however, the relative efficacy of the procedure should be judged
on the stricter standard (272, 366, 368, 376). Follow-up results based
only on participants who respond or who are readily available are
especially suspect (24, 272, 366, 368, 376).

The final issue that commonly affects outcome data from smoking-
modification studies involves the replicability and generalization of
results. Programs and studies with reportedly very similar procedures
have produced highly variable patterns of results (24, 26, 40, 171, 200,
226, 230, 366, 376, 413). This, it seems, is due in part to the variability
introduced by small samples and population differences (24, 171, 226,
272) and the inadequacies of theoretical models guiding the descrip-
tions of treatment variables (24, 272, 306, 398). In an effort to minimize
these deficiencies, the NICSH Guidelines (272) stress the need to
describe completely the recruitment and selection of participants, their
characteristics, and the specifics of each aspect of treatment. Keutzer,
et al. (200) have also discussed the problems of uncontrolled variability
from group treatment and inexperience of the therapist or experi-
menter.

Thus, conclusions regarding the relative efficacy of treatments can
be reliably made only when methodological deficiencies are at a
minimum (272). The quality of the data has improved markedly since
the early reviews (24, 200, 366), but almost all studies remain deficient
in some respect (368, 376). Many programs have collected little or no
objective follow-up data, and the lack of methodological rigor
compromises the results of many others that have. Therefore, based

19—8
upon current data, the replicability and general utility of almost all
procedures can be only tentatively assessed.

Review of General, Nonspecific Interventions

A variety of interventions has been developed and offered with the
primary goal of aiding a group of smokers to become nonsmokers
rather than testing how the procedures may work (398). Various
reviewers have analyzed the data on this type of intervention, which
includes public service and proprietary withdrawal clinics, individual or
medical counseling, and large scale coronary prevention trials. Except
for the coronary prevention trials, the clinical-treatment focus of these
interventions has resulted in multiple uncontrolled clinical replications,
often without adequate outcome data (24, 40, 171, 200, 245, 366, 368,
376). Additionally, the vast public health campaign of recent years
should be considered as a special class of general, nonspecific
interventions both to prevent smoking onset and to stimulate cessation
(24, 40, 200).

Public Health Educational Campaigns

The public health campaign against cigarettes has produced notable
changes in public awareness of the health consequences of cigarette
smoking (175, 269, 271, 422). It appears that the dramatic changes
noted in adult smoking, especially among middle-aged males and
certain professional groups (86, 100, 121, 271, 421), can be attributed
largely to the effectiveness of information and educational campaigns
since 1964 (130, 270). Moreover, Warner (428) has estimated that the
effect of specific “events,” such as the 1964 Surgeon General’s Report,
on cigarette consumption (mean number of cigarettes consumed per
day) may appear small and transitory, but that the cumulative effect
of persistent publicity appears to have reduced consumption by 20 to 30
percent below its predicted 1975 level.

More specifically, O’Keefe (284), in a study on the impact of
television anti-smoking commercials during the late 1960’s, revealed
changes in attitudes and reported reductions in consumption but little
direct impact on smoking cessation. Forty-two percent of those
motivated to quit felt the commercials acted as an incentive, but only 1
percent of the ex-smokers credited the commercials with helping them
quit. Similar minor effects were noted in a smaller trial with anti-
smoking posters (5). Ryan (353) reported the results of an entire
community’s attempt to quit in 1970. Thirty-seven percent of the
adults attempted to quit, and 14.2 percent of the males and 3.9 percent
of the females were still reporting abstinence 7 months later, with
higher socioeconomic groups being more successful. The Avdel
smoking project (98) also seemed to have produced small but
meaningful changes in both smoking attitudes and behavior with a

19—9
worksite campaign. These specific and general results of the public
health campaigns appear very similar to other British (343) and
worldwide experiences (130, 301).

Public Service and Proprietary Clinics

It is interesting to note that Bernstein’s (24) comment that the
educational campaigns have affected research and clinical activities
more than smoking behavior still seems valid. Public service and
proprietary programs have proliferated since 1964. Schwartz and Rider
(376) have provided a summary of the published and unpublished data
on these types of programs. Many such smoking-withdrawal clinics
offered by voluntary agencies have been intermittent and rarely
evaluated. The group program of the American Cancer Society (ACS)
(2, 3, 160) and the 5-Day Plans of the Church of the Seventh Day
Adventists (252, 258, 254) have, however, remained very active in
providing public service treatments to smokers. Unfortunately, while
the two programs together have probably helped more smokers than
any other organized effort (245, 368, 376), only limited published
outcome data are available for consideration.

The 5-Day Plan has become standardized and involves five
consecutive 1!/2- to 2-hour sessions focusing on immediate cessation,
and dietary, physical, and attitudinal changes to reduce withdrawal
effects (252, 254). Because of its clinical focus, almost all evaluations
have been without controls (117, 146, 147, 148, 218, 252, 258, 254, 267,
298, 366, 376, 403, 412), with gocd immediate abstinence rates of
approximately 60 to 80 percent, but with an approximately 50 percent
relapse by 1- to 3-months post-treatment. Unfortunately, clinical
claims of abstinence among 33 to 40 percent of participants beyond a
year post-treatment (146, 147, 148, 253) are markedly discrepant from
other clinical demonstrations (213, 267, 298, 361, 412). Guilford’s
comparative study of the 5-Day Plan (137, 138) found abstinence rates
of 16 to 20 percent at 1 year that may not differ from unaided attempts
(137, 188, 412). Nevertheless, the program appeared to be more
successful with males (137, 138, 267, 403) and when higher expectation
of success was reported by participants (361). Results of all studies are
based on unverified self-reports, often only from subjects completing
all treatments (366, 376).

Available long-term abstinence outcome data on the ACS group
programs (2, 3) also appear to be somewhat disappointing. The one
available evaluation of the ACS groups, which focus on insight
development, group support, and self-selected cessation techniques,
was conducted on 29 clinics in Los Angeles from 1970 to 1973 (318).
Telephone follow-ups were completed on 354 subjects selected from a
random sample of 487 of the original 944 participants. Abstinence rates
based on the total random sample were 41.7 percent at post-treatment,
and 30 percent at 6-month, 22 percent at 12-month, and 18 percent at

19—10
18-month follow-up points (245, 318, 378). In the subsample group of
354 subjects who were contacted (318), 28.4 percent of the males and
20.3 percent of the females reported abstinence.

Other clinics with similar or more elaborate formats have reported
fairly equivalent outcome data (63, 81, 82, 114, 158, 178, 213, 274, 286,
289, 433, 488, 440, 448). The Smoking Withdrawal Study Centre in
Toronto (81, 82, 378) used comprehensive educational groups with 472
smokers and obtained successful abstinence in 28.6 percent of all
participants at 1-year follow-up, with 33.9 percent of the men and 20.8
percent of the women being successful. However, carboxyhemoglobin
(COHb) assessments revealed that 22 of the 107 (20.6 percent) reported
ex-smokers had levels over 5 percent, which strongly suggested
smoking. A 5 percent quit rate was noted among a no-treatment
control group. In a population based sample, Isacsson and Janzon (178)
were able to produce abstinence during an intensive 6-week program
among 31 of 51 participants (60 percent), with 17 (83 percent)
remaining nonsmokers at 8- to 9-month follow-up. Abstinence was
verified by COHb determinations. West and his colleagues (433)
followed up 559 smoking-cessation clinic participants 5 years later and
found 17.8 percent of the contacted sample reporting abstinence.
Approximately two-thirds of those who had quit during the clinic had
returned to smoking, while only 8 percent of the unsuccessful
participants were reporting abstinence at follow-up. Older males who
had lighter smoking habits and more stable environments appeared to
be most successful. Research clinics (to be discussed in more detail
elsewhere in this report), offering similar treatment formats, have
reported similar 15 to 20 percent long-term abstinence among
participants (341, 378, 374, 380, 381, 382).

In light of these data on public service and research withdrawal
groups and clinics, the claims of more impressive results by proprietary
programs must be viewed with caution (116, 245). Schwartz and Rider
(376) reviewed a variety of unpublished data on commercial methods,
but only one published evaluation of a commercial method is currently
available. In this study (194), records of 553 participants of the
SmokEnders program in 1971 were examined and a 31/2 to’4-year
follow-up was attempted on the 385 (70 percent) who were not smoking
at treatment termination. Only 167 (43.4 percent) were contacted; of
these, 57 percent of the males and 30 percent of the females were not
smoking. Schwartz and Rider (376) noted, however, that, even if the
smoking rates of those contacted at follow-up accurately represent the
total successful sample, the long-term success based on all participants
(including treatment dropouts) would be about 27 percent rather than
the reported 39 percent. As the men and women were reported to have
been about equally successful at treatment termination, the higher
follow-up success rate for males would still seem valid.

19—11
In viewing the data from many ciinics relative to the 16 to 19 percent
success at 1-year follow-up noted in Guilford’s (137, 138) and Schwartz
and Dubitzky’s (373, 374) unaided control groups, the impact of many
programs appears to have been minimal. Bernstein’s (24) conclusion
still seems valid: clinics can serve a very useful purpose when more
effective modification techniques are developed for general distribu-
tion, but uncontrolled use of nonvalidated notions cannot refine those
procedures. The attempts to analyze more carefully the clinic format
has produced some enlightening data (81, 82, 187, 138, 178, 318, 341,
361, 373, 374, 380, 381, 382, 433). Long-term results imply that males in
these clinics fare better than females during maintenance (81, 82, 137,
138, 267, 341, 376, 403, 433). Moreover, the comprehensive follow-up
and physiological validating of some studies (81, 82, 178, 373, 374)
highlight how misleading early success based on self-reports can be.
The placebo effect noted in control groups highlights the fact that
many of the treatment effects of clinics remain undefined (373, 374).
More effort should be made, therefore, to evaluate on-going clinical
activities so that researchable hypotheses can be illuminated for
further controlled study (24, 394).

Individual and Medical Counseling

Smoking-cessation counseling by professionals in private practice is
known to exist, but published data on its efficacy are very rare. A
report on two psychotherapist-led groups suggests that long-term
therapy may help some smokers (39); however, the cost of such
treatment would seem prohibitive (245). In controlled studies of the
type of individual and group counseling formats that could be easily
and less expensively disseminated, Schwartz and Dubitzky (373, 374)
and the American Health Foundation (380, 381, 382) produced 1-year
abstinence rates ranging from 13 to 30 percent with no clear
superiority for individual or group therapy. While individual counsel-
ing styles seemed to affect initial success and dropout rates, there were
no differences in effectiveness during follow-up (186, 431).

Since smokers have become almost uniformly aware of the health
risks of smoking (269, 271, 422), they view the physician as an
important person in the quit-smoking decision (271). However, only
about 25 percent of smokers surveyed in a national telephone interview
reported having been advised by their physician to quit (271). Almost
all physicians are convinced of the health consequences of smoking and
have made dramatic change: in their own smoking (121, 421), but many
seem reluctant to confront their smoking patients until serious effects
are present (55, 338). Nevertheless, numerous studies of ex-smokers
have shown that linking the increase of symptoms, such as coughing or
breathlessness, to smoking was a major precipitant for unaided
quitting (51, 128, 150, 152, 190, 294, 389, 390, 399, 400, 418, 419).

19—12
Rose (338) and Lichtenstein and Danaher (227) have reviewed the
issue of physician counseling and its efficacy. In general, it appears
that physicians have heen discouraged from this role (338) and are
effective as counselors only when dramatic symptoms are present (227,
338). Several uncontrolled studies, done primarily in England, have
shown varying success. Early studies in this country showed minimal
effects (244, 322). Studies abroad, on the other hand, have evaluated
several important aspects of the process. Porter and McCullough (312)
produced only 5 percent abstinence at 6 months in a briefly-counseled
group, while 4 percent quit in a randomly defined uncounseled group.
Handel (153) reported more impressive results from one brief session
with 17 of 45 (88 percent) males and 6 of 55 (11 percent) females
reporting abstinence at l-year follow-up. When patients presented
current respiratory symptoms, Williams (443) and Burns (51) found a
higher response to brief counseling. Burns (51) reported 35 of 66 (58
percent) males and 9 of 28 (32 percent) females reporting completely
stopping 3 months after the visit. Similarly, Williams (443) found that,
of 204 patients routinely counseled, 59 of the 160 (37 percent) who
could be contacted at 6-month follow-up were reporting abstinence,
with males and females being about equally receptive.

Some of the variability of response may be due to individual
physician styles. Pincherle and Wright (302) followed up a total of
1,493 business executive smokers for 1 to 2 years after a regular
physical where smoking-cessation advice was given. Thirteen percent
reported quitting and 11 percent indicated a reduction in rate of 30
percent or more; however, when the results were analyzed across
various physicians giving the message, success (quitting or 30+
percent reduction) rates varied from 35 percent to 17 percent. In a
similar follow-up of antismoking advice given during annual physicals,
Richmond found 118 of 543 (22 percent) quit for at least 1 vear; 15
subsequently relapsed, leaving a long-term success rate of 19 percent
(329). Unfortunately, no physician-counseling study has utilized
techniques to validate self-reported behavior change.

Considering the brief nature of the contact and the lack of specific
maintenance follow-up, the reported rates of abstinence seem encour-
aging. A study by Raw (319) has suggested that both a physician’s
message and counseling by a health professional in a white coat were
-mportant in producing cessation, also suggesting that health profes-
sionals other than physicians should become more involved. Peabody
291) reported that with a well-developed program, 25 percent of
smokers will quit after the initial counseling, 25 percent will quit after
several attempts, 20 percent will eventually stop with difficulty, and
only 30 percent will never respond. These expectations may be high for
1 general patient population, but cessation data on special groups of
yatients with current medical problems related to smoking are
ncouraging.

19—13
Patients hospitalized with their first myocardial infarction (MI)
provide a dramatie example of this. Thirty to fifty percent of the
smokers in this group permanently stop smoking after only routine
advice (4, 1], 68, 157, 338, 430, 432, 442). Follow-ups on hundreds of
such patients reveal that relapses back to smoking are uncommon, with
50 percent quit rates often maintained for 1 or more years (11, 68, 338,
430, 432). When more intensive counseling and active follow-up
support were undertaken in a study by Burt and associates (52), 70 of
114 (61 percent) of cigarette smokers and 9 of 11 (82 percent) of cigar
and pipe smokers stopped smoking after hospitalization, and only 19
(15 percent) of the smokers made no changes. At the 1-year follow-up,
9 of the immediate quit group (11 percent) and 13 of 22 (59 percent)
who quit later relapsed, leaving 79 of 125 smoking (cigarette, pipe, or
cigar) patients reporting abstinence (63.2 percent) with 27 (21.6
percent) having reduced. Among 120 patients given conventional
advice and not followed up in the special clinic, only 27 of 98 (27.5
percent) of the smokers were reporting abstinence and 27 (27.5
percent) reporting reduction at the 1-year follow-up.

Thus, physicians and other health professionals have great opportu-
nities for anti-smoking counseling. Both Rose (338) and Lichtenstein
and Danaher (227) warn, however, that the private practitioner should
avoid unrealistic expectations and underestimations of the time
required. Various guidelines have been offered on the office manage-
ment of cigarette smoking (113, 115, 166, 291, 307, 309, 402);
Lichtenstein and Danaher (227) provide a comprehensive format and
suggestions. Clearly, health care professionals can play a dramatic role
by being nonsmoking models, by linking current symptoms to smoking,
and by aiding smokers in the decision to quit alone or with additional
help. But as Rose (338) and Lichtenstein and Danaher (227) have
pointed out, additional research is needed to test techniques applicable
for office-guided cessation programs.

Large-Scale Coronary Prevention Trials

Middle-aged men judged at risk but not exhibiting coronary heart
disease (CHD) provide a special challenge for smoking counseling (336,
337). Since cigarette smoking together with serum cholesterol and
blood pressure levels are considered the major risk factors for CHD (36,
420), preventive trials have attempted to reduce the incidence of CHD
in study samples by using a multifactor approach. The Coronary
Prevention Evaluation Program (391, 392) was an initial 7-year
feasibility test of this approach among 519 coronary-prone men aged
40 to 59 at intake. Only 116 of the original 191 smokers remained active
in the study, and more emphasis was given to nutritional counseling
than to smoking counseling. Nevertheless, 43 of the 116 (37.1 percent)
Subsequently, other trials were initiated in Europe (449). Wilhelm-
sen (439) established a comprehensive cessation program for use in a
field trial in Sweden (441), but long-term results are not available. In a
controlled trial of the effects of anti-smoking advice among 1,470
ecoronary-prone London civil servants (324), 51 percent of the 714
randomly assigned to anti-smoking clinics stopped smoking by the end
of 1 year. Only 31 percent were reporting complete abstinence, as
many converted to pipes and cigars (338). In general, the preliminary
results of the European multifactor prevention trials are only
moderately successful, with abstinence in 16 to 28 percent of the
smokers after 1 year (449).

In 1972 the Multiple Risk Factor Intervention Trial (MRFIT) was
initiated in this country (265, 266). One of the largest and most
ambitious of the multicomponent efforts to influence cigarette
smoking behavior among middle-aged men, this smoking intervention
attempt is occurring within a broad 6-year coronary prevention
program also intended to reduce serum cholesterol and blood pressure
levels in over 6,000 men aged 35 to 57 at increased risk of coronary
disease (410). Initial intense intervention involving multicomponent
group or individual sessions produced abstinence in approximately 43
percent of the smokers by the first annual examination (280).
Biochemical assessments are being made to validate the self-report
data. Continued intervention and maintenance contacts have produced
successful cessation in other participants who had not formerly quit
and in participants who had returned to smoking (280).

Two studies have focused on total populations rather than selected
high-risk groups. The North Karelia Project (204, 31 6) has been
providing a comprehensive community program since 1972 to reduce
the very high rate of cardiovascular disease in eastern Finland. By the
end of the first year of intervention, the proportion of males aged 25 to
59 in the North Karelia district who smoked decreased from 54 percent
to 48 percent, while female smoking rates have remained at about 11 to
13 percent throughout the 5 years of treatment. These encouraging
ehanges in male smoking behavior were maintained, with the 5-year
follow-up survey reporting 42 percent of the adult men still smoking.

More specific data are available on the field study conducted by the
Stanford Heart Disease Prevention Program. An extensive 2-year,
mass-media campaign (234) was presented to two California communi-
ties to persuade the general public to modify eating and smoking
behaviors in order to reduce cardiovascular risk. A third community
served as control (101, 235). Face-to-face behavioral counseling (101,
247, 258) was offered to two-thirds of the high-risk subjects in one of
the media communities. Three years after the program started, the
proportion of smokers had decreased by 3 percent in the control
community, by 8 percent in the media-only community, and by 24
percent in the media-plus-counseling communities (101, 248, 259). Fifty

19—15
percent of the high-risk smokers receiving face-to-face counseling, but
only 11 percent receiving just media, had quit (101, 248, 259).
Thiocyanate monitoring was performed to validate self-reports.

When the risks of smoking are made more immediate and salient,
and both skills and support to change are provided, meaningful
reductions are possible. The multifactor trials reveal that when
smokers are sufficiently educated regarding their risks, they respond
much like the post-MI patient and quit immediately and relapse less
than would be predicted. The most successful multifactor trials have
involved expensive face-to-face intervention techniques and extensive
follow-up contacts (280, 410) or costly and well-conceived behavioral
and media programs (101, 204, 235, 247, 316). Hence, more work is
needed to translate the skills developed from these research trials into
office practice and public health campaigns (227, 338). It should be
noted that the effective programs involved face-to-face intervention
techniques which were both intensive and expensive.

Controlled Experimental Research on Intervention Strategies

A wealth of research data relevant to the modification of smoking
behavior has been produced. Early controlled research tended to
produce unimpressive results (24, 200, 366). Schwartz and Dubitzky
(373, 374) conducted an exemplary study of what appeared to be the
best treatment options available in the late 1960’s (24, 200, 366). Initial
results suggested that group or individual therapy had moderate
effects on smoking; but, by the end of a 1-year follow-up, not one of
the seven experimental conditions was superior to the no-contact or
minimal-contact controls (373, 374). Recent progress has begun to
highlight both what strategies may be more effective and why they
may work. Because these data have been comprehensively evaluated
and discussed in recent reviews (26, 29, 226, 245, 868, 376), this section
will emphasize primarily the major trends in this research history.

Drug Treatments

The psychopharmacology of smoking and its relationship to smoking
behavior and cessation are discussed in some length elsewhere in this
report and in recent reviews (46, 136, 181, 183, 349). While research
(349, 359, 360) continues to suggest that there are pharmacological
determinants for smoking, the identification of chemical agents either
to substitute for smoking or to minimize withdrawal symptoms has
been frustrating and difficult (136, 181, 183).

Early research on Lobeline as a nicotine substitute was equivocal (24,
200, 366). The utilization of the substitute in a clinic format seemed to
at least enhance short-term effectiveness (93, 341), but the double-
blind study by Davison and Rosen (77) indicated that Lobeline was no
more effective than an appropriate placebo. More recently, a nicotine

19-16
chewing gum has been developed and tested as a cessation aid (47, 102,
103). Double-blind studies using the gum in cessation clinics suggested
that it is significantly more effective than placebos (41, 185, 283, 352),
but, beyond the control of withdrawal symptoms (364), its effects
appeared to be a small component in the overall success (352).

Combinations of drugs to reduce withdrawal symptoms have been
used in various clinics (180, 341, 438, 440); however, the double-blind
study by Schwartz and Dubitzky (373, 374) of meprobamate with and
without individual or group therapy suggested that the placebo, if
anything, was more effective. While all treatment conditions were
initially superior to questionnaire and screened no-treatment controls,
the prescription-only and prescription-plus-individual-counseling had
lower (8.3 percent and 18.9 percent) abstinence rates at 1-year follow-
up than the controls (16.7 and 19.4 percent) (373, 374).

Other chemicals have been tested in Europe with some initial success
(136, 363), but additional evaluations are needed (136, 376). Rosenberg
(340) reported initial success in reducing consumption in a double-blind
study of an antismoking chewing gum that caused an unpleasant taste
when tobacco was subsequently smoked. The gum’s efficacy as a
cessation aid was not tested. Current data suggest that the usefulness
of pharmacological cessation aids has yet to be unequivocally
demonstrated. While aids such as nicotine gum may be useful in the
control of withdrawal symptoms in some smokers, current research
suggests that they would need to be combined within a broader
program to produce and maintain abstinence (136, 352).

Hypnosis

Clinicians have claimed from 42 to 86 percent of their clients treated
with hypnotherapy were abstinent at 6- to 12-month follow-up (66, 67,
148, 278, 358, 395, 429, 450). Unfortunately, these claims have not been
substantiated in controlled research. The early research was chaotic
and methodologically poor, leading Johnston and Donoghue (189) to
conclude that “there is almost no good research evidence attesting to
. the effectiveness of hypnosis in the elimination of smoking behavior”
(p. 265). Moreover, Spiegel, a leading proponent of self-hypnosis,
claimed that the actual success rate may be closer to 20 percent long-
term abstinence (87, 388). Orne (285) considered both the theoretical
foundations and research data for hypnosis and concluded that its
effects can best be categorized as a placebo response which leads to
nontraumatic cessation through both the mystique of the procedure
and the hypnotic suggestions.

The data from several recent studies do not refute these conclusions.
Pederson and associates (295) found that 9 out of 16 (54.3 percent) of
the subjects in a hypnosis-plus-counseling group were reporting
abstinence at 10-month follow-up as compared to 12.5 percent for
counseling-only or waiting-list control groups. As there was only 8

19—17
percent abstinence for a group treated with hypnosis only, they
concluded that hypnosis can enhance the effects of group counseling;
alone, it may be insufficient as a cessation procedure. When Shewchuk
and associates (382) allowed smokers attending clinics to choose group
therapy, individual therapy, or hypnosis, 193 of 571 (34 percent) chose
hypnosis. The group therapy-reported abstinence rate (49 percent) was
significantly superior to those of both hypnosis (88 percent) and
individual counseling (33 percent) at treatment termination. By 1-year
follow-up, however, all three conditions showed marked relapse,
leaving only 17 to 21 percent of the participants reporting abstinence.
While assignment to conditions was self-selected and nonrandom, the
failure of hypnosis to replicate clinical claims remains important.

Barkley and associates (18) found that group hypnosis did not
significantly differ from an attention-placebo control in mean smoking
rates at any point during treatment or follow-up, but it had more
subjects claiming abstinence at the 12-week follow-up point (4 of 8 vs.
1 of 9). At the 9-month follow-up, only two of eight (25 percent) of the
hypnosis subjects were reporting abstinence versus none for the
control. Francisco’s (105) unpublished dissertation appeared to have
reached a similar conclusion. It has been suggested that a 15 to 20
percent success rate for hypnosis may reflect the expected proportion
of subjects highly susceptible to hypnosis (297).

Social Psychological Approaches —

Higbee (159), Leventhal (216, 217, 218, 219), and Rogers (332) have
reviewed most of the data from field and laboratory studies conducted
to test responsiveness to persuasive communication regarding ciga-
rette smoking. While most studies on smoking have produced attitude
changes without marked or lasting reductions in smoking behavior
(181, 182, 231, 239, 244, 303, 321, 401), this area of research has clarified
several basic aspects of the smoking cessation process. The results and
implications of these studies have been summarized by Leventhal (216,
217, 218, 219) and Rogers (332).

Janis and Hoffman (282) demonstrated the facilitating effects of
daily telephone contacts that persisted well into follow-up despite
termination of the contacts. Unfortunately, mean-rate reductions
rather than abstinence rates were reported. Rogers and associates (333,
334) have recently documented the long-term impact of several
communication strategies on smoking behavior. They reported signifi-
cantly higher abstinence for high-fear versus low-fear messages in a
coliege sample at 3-month follow-up (22 percent vs. 7 percent), and ina
community sample at 1-year follow-up (18.8 percent vs. 0 percent).

Suedfeld’s unexpected results with a single exposure to 24-hour
sensory deprivation (SD) are also impressive (405, 406, 407). In a pilot
study with five subjects, four quit after treatment and were reporting
abstinence for 1 to 3 months afterwards (406). In a controlled study

19—18
(407), almost all SD subjects were reported to be abstinent at
treatment termination, and 10 of 37 (27 percent) appeared to remain so
at 12-month follow-ups when only 4 of 35 (11.4 percent) of control-
condition subjects were reporting abstinence. Recently, Suedfeld and
Best (405) piloted a combination of SD with a complex behavioral
program involving aversive smoking and reported abstinence in four of
five subjects for over 8 months.

This latter finding is supportive of Leventhal’s (216, 219) conclusion
that attitude change without a meaningful plan for action will not
produce behavioral change. Hence, additional] integrations of attitude
and behavior change procedures seem worthy of investigation.

Social Learning and Behavior Modification Approaches

Research based on experimental and social learning theories (12, 14,
106, 168, 169, 172) has produced a wide diversity of controlled studies.
Unfortunately, most of the early research on techniques that had been
successful with other behavioral problems (106) or were derived from
the principles of experimental psychology and laboratory research on
behavior change proved to be minimally effective in producing long-
term changes in smoking behavior. While early reviewers (24, 200, 230)
acknowledged these discouraging initial treatment results, they
concluded that the more empirical approach of these procedures made
them the most promising. These hopes have been only partially
fulfilled (243, 451).

Specifically, many studies have been more concerned with theoreti-
eal comparisons based upon evaluations of smoking-rate changes than
with developing techniques with documented efficacy based on long-
term abstinence data. Techniques were often found to be at least
temporarily superior to control conditions, but the effects either
vanished during follow-up or no meaningful follow-up was conducted
(25, 58, 59, 64, 70, 107, 132, 135, 139, 155, 197, 199, 201, 206, 207, 209, 212,
215, 220, 221, 242, 255, 260, 278, 276, 280, 281, 287, 317, 877, 384, 394, 408,
409, 426, 434, 435, 436, 437, 447).

This pattern has been especially common in dissertation research on
smoking. Most such dissertation research has been conducted by
‘doctoral candidates and supervised by committees who generally have
solid experimental and methodological backgrounds but limited clinical
experience with smokers (225). Armchair and theoretical analyses of
smoking have too often led to experimental and control conditions of
some theoretical interest but which typically produced no relative
differences among groups at follow-up and weak absolute results as
measured by abstinence rates (225, 376). Furthermore, graduation
pressures usually lead to insufficient follow-ups of only 1 to 3 months
(225). The number of unpublished doctoral dissertations of this type
document how much well-meaning effort has been devoted to the
production of largely inconclusive results (10, 20, 34, 35, 38, 60, 69, 87,

19—19
88, 96, 118, 123, 125, 127, 134, 146, 161, 187, 188, 191, 196, 236, 249, 268,
277, 292, 315, 328, 342, 357, 865, 385, 886, 411).

Overall, the methodology of the research based on learning-theory
approaches has been improving (26, 226, 376). Most studies have
utilized appropriate designs and controls, follow-ups are becoming
longer, and, most encouraging, validation of self-reported abstinence
has become more common. Confirmations by informants in the
participant’s natural environment have been the mainstay (8, 21, 22,
27, 28, 31, 32, 59, 64, 71, 85, 123, 141, 142, 197, 202, 206, 210, 229, 240, 242,
251, 279, 292, 318, 362, 394, 446). However, carbon monoxide monitoring
(71, 206, 351), threatened or actual urine nicotine analyses (308, 409), a
bogus marketing survey procedure (94), and attempted (80) or actual
(48, 246) thiocyanate analyses have now been reported. Although the
outcome data on most procedures have been quite variable, the stricter
methodology of these studies has encouraged continued refinement of
interventions. More recently, effective multicomponent programs have
begun to develop from this earlier research. The wealth of studies will
be discussed briefly, therefore, with special emphasis given to those
research trends that have produced programs with documented
effectiveness. More detailed discussions of the literature are available
in past (24, 200, 230, 366) and recent (26, 29, 226, 245, 368, 376, 413)
reviews.

The research in this area can be grouped loosely into two broad, but
not mutually exclusive, categories: (1) behavioral self-control strate-
gies utilizing high participant involvement and (2) aversion strategies
designed to reduce the probability of the smoking response (226).
However, the most effective programs have tended to be multicompo-
nent interventions which combine certain strategies from both
categories.

Self-Control Strategies
Stimulus Control

The basic philosophy of behavioral self-control treatments has been to
provide the subject first with increased awareness of the target
behavior and controlling stimuli and then with specific self-manage-
ment skills to control the target behavior (13, 14, 198, 241, 314, 414,
415). Therefore, self-monitoring of individual smoking behaviors has
been a fundamental element in all behavioral self-control programs. As
a sole treatment, self-monitoring has rarely produced more than
temporary treatment effects (60, 87, 109, 25U, 251, 288, 365, 411) and
has been classed with the nonspecific treatment factors common to
almost all behavioral programs (251). Self-monitoring has usually been
combined within stimulus control treatments to make subjects aware
of the specific environmental and internal cues associated with
smoking urges and behaviors.

19—20
‘These stimulus control programs have been based on learning-theory
formulations (168, 169, 172) of smoking behavior that suggested
cessation is difficult because smoking is prompted by such a variety
and range of cues. Subjects were taught to reduce the strength of
these cues either by eliminating smoking from an increasing number of
situations or by making time intervals the only controlling cue (24, 26,.
226).

While this process theoretically should, with rare exceptions (311,
344, 345), make cessation easier, most subjects were reported to have
difficulty reducing below 10 to 12 cigarettes per day (8, 10, 23, 59,.104,
139, 221, 242, 313, 377). It has been suggested that, when most smokers |
reached. that reduced level, each cigarette became more reinforcing
and difficult to give up (104, 243).

Most studies involving a variety of stimulus control and other self-
management techniques were shown to be at best only temporarily
superior to control conditions. These studies have produced, in general,
the common pattern of temporary reduction but rapid relapse and
long-term abstinence rates that did not differ from those expected
from nonspecific treatments (10, 23, 60, 69, 87, 104, 125, 182, 189, 146,
155, 188, 191, 196, 197, 199, 221, 242, 260, 264, 273, 277, 279, 280, 328, 355,
365, 377, 385, 386, 411, 435). Even when applied within more complex,
multicomponent programs, the stimulus control-based treatments
often produced only moderately encouraging findings (48, 104, 155,
255, 273). Some encouraging applications have been noted (44, 45, 308,
416), however, especially when the programs develop from systematic
research and the programs offer behavioral training in a wide range of
skills (42, 310).

. Contingency Contracting

One specific technique that has produced some encouraging data
involves the depositing of money for later disbursement based on
attainment of specified goals. Early research on the technique was
equivocal (24, 200, 224, 230), but several studies have produced
impressive results. Elliot and Tighe (95) reported 84 percent abstinence
at treatment termination, with 4 of 11 (36 percent) in two other groups
followed up 15 to 17 months after treatment. However, the treatment
also involved public pledges, stimulus control techniques, and group
support.

Winett (446) found that 50 percent of the subjects in contingent
repayment condition were abstinent, validated by informant reports,
at 6-month follow-up, but only 23.5 percent of those in noncontingent
repayment were abstinent. Multiple case studies by Axelrod and
associates (6) and a study by Rovner (342) were also encouraging.
Brengelmann (44, 45) has reported notable success in recent studies
utilizing contingency contracting within a treatment-by-mail program.
Forty-seven percent of those responding to the 15-month follow-up

19—21